All Questions
Tagged with performance storage
90
questions
4
votes
2
answers
971
views
Windows File server heavy cpu usage - Render Farm with many clients reading data at once
I have a windows file sever with Dual Xeon E5-2650v4 cpus. They are 14 core each so 28 cores total.
The network interface is a Mellanox Connectx3 40gbs.
The disk drives in the server are 7.68TB Sata ...
- 41
0
votes
1
answer
418
views
High util iostat
I have server with oracle db. It works on Dell Unity storage.
The server has very slow disk IO.
How can I fix it?
Here are some details of my server:
OS - OL8
Kernel - 5.4.17 uek
Storage - Dell ...
- 1
0
votes
2
answers
870
views
Disk iops vs latency
I am trying to understand the diference between iops and latency, this is my question:
If an hypothetical disk had 2 io per second, then its latency would be 0.5 seconds?
2
votes
2
answers
670
views
High speed network writes with large capacity storage
I have a NAS running Samba with a 20T ZFS pool with one raid1 vdev with two spinning rust drives. I have 16G RAM in the machine right now. The storage is used for continuously growing, permanent ...
- 394
1
vote
1
answer
776
views
MegaRAID: performance difference between volumes
I have a MegaRAID SAS 9361-8i with 2x 240GB SATA 6Gbps SSD in RAID1, 4x 10TB SAS 12Gbps HDD in RAID6 and 4x 480GB SATA 6 Gbps SSD in RAID5:
-------------------------------------------------------------...
- 3,327
3
votes
0
answers
2k
views
Gluster Performance
currently I try to setup a Gluster cluster and the performance is strange and I'm not sure, if I configured something wron. I'm using 4x Hetzner root server running Debian Buster with Intel i7, 128GB ...
- 508
26
votes
4
answers
7k
views
Is it reasonable to use NFS on a production web server?
Can NFS be reasonably used on production servers as a means of connecting a compute server to a storage server, assuming the connection is over a LAN 1Gbe or 10Gbe connection?
There's obviously some ...
- 371
2
votes
1
answer
104
views
MYSQL vs. InnoDB [closed]
mysqltuner output.
-------- Storage Engine Statistics -----------------------------------------------------------------
[--] Status: +ARCHIVE +BLACKHOLE +CSV -FEDERATED +InnoDB +MEMORY +MRG_MYISAM +...
4
votes
1
answer
2k
views
Slow RAID performance for Samsung EVO's
So I've been suspecting performance bottlenecks on a Samsung EVO 850 RAID1 for sometime, but honestly been too lazy to look into it. Now I was starting a new home project, involving a Vmware ESXi host ...
- 51
0
votes
1
answer
12k
views
What's the difference between 10k/15k SAS HDDs and SATA SSDs?
What is the rule of thumb in performance characteristics and differences between 10k / 15k revolutions per minute (RPM) Serial Attached Storage (SAS) hard disc drives (HDDs), versus Serial ATA (SATA) ...
- 14k
2
votes
1
answer
507
views
12G SAS Expander with 6G disks limiting speed?
I'm building a server with 24 SATA III disks connected to a Chenbro 12G SAS expander on a 12G SAS HBA (LSI 9300) with a single SFF-8643 cable. Aggregate write speed is slower than I hoped for, and I'...
- 200
10
votes
1
answer
2k
views
Slow sequential speeds on 9x7-drive raidz2 (ZFS ZoL 0.8.1)
I'm running a large ZFS pool built for 256K+ request size sequential reads and writes via iSCSI (for backups) on Ubuntu 18.04. Given the need for high throughput and space efficiency, and less need ...
- 261
7
votes
3
answers
2k
views
3Gbps SSDs in RAID1 VS 6Gbps SAS Drives in RAID1 - Performance Difference?
I have bought two Intel DC S4500 480 SSDs (comes with 5 year warranty) to use with a HP G7 DL360 1U Server in RAID1. Unfortunately after purchasing the disks, I figured out that the integrated P410i ...
- 235
4
votes
1
answer
7k
views
Optimal ARC and L2ARC settings for purpose specific storage application
I am configuring a server that runs 3 ZFS pools, 2 of which are rather purpose specific and I feel like the default recommendations are simply not optimized for them. Networking is facilitated by dual ...
- 163
0
votes
1
answer
452
views
Real Time Burst IO Balance on EC2 Instance
I'm trying to troubleshoot some disk performance issues on an ec2 instance using a 60 GiB gp2 volume. The cloud watch metrics don't seem very fine grained. Is there a way to view in real time the ...
- 103
14
votes
2
answers
13k
views
Impact of RAID levels on IOPS [closed]
With regard to IOPS, I have seen several sources on the web that suggest the IOPS of a given number of disks is simply the IOPS of a single disk multiplied by the number of disks.
If my understanding ...
- 1,852
7
votes
1
answer
4k
views
Better performance when HDD write cache is disabled? (HGST Ultrastar 7K6000 and Media Cache behavior)
Attention, please. Long read.
During initial performance tests of Hitachi Ultrastar 7K6000 drives that I'm planning to use in my Ceph setup I've noticed a strange thing: write performance is better ...
- 91
12
votes
1
answer
9k
views
What actually happens when I mark a disk as SSD in ESXi?
In VMWare ESXi / vSphere, you can mark a disk or LUN as "Flash", indicating that the disk is an SSD (or the LUN is SSD/flash storage backed)
When you do, you get the following warning:
Marking HDD ...
- 9,198
34
votes
8
answers
13k
views
Why do IOPS matter?
I understand what IOPS and throughput are. Throughput measures data flow as MB/s and IOPS says how many I/O operations are happening per second.
What I don't understand is why many storage services ...
- 513
2
votes
1
answer
241
views
How do you store VMs on HDDs with data integrity AND performance?
First off I know this topic easily gets subjective, but I'm trying to avoid that, as there should be at least 1 good answer to this in a sea of bad answers and it's hard to find.
At first sight my ...
- 389
5
votes
1
answer
134
views
storage for virtualization: 10x15k sas w/ 2gb cache OR 10x10k sas w/ 4gb cache
sorry for the long title.
this is the point: due to a number of decisions it doesn't worth to share here and which do not completely belong to me, I'm facing the choice of 2 different storage ...
- 654
31
votes
8
answers
10k
views
Write speed requirement : 1.1GB/s possibilities?
We will have a machine at work, that on peak performance, should be able to push 50 ("write heads") x 75GB of data per hour. That's peak performance of ~1100MB/s write speed. To get that from the ...
- 749
1
vote
1
answer
1k
views
IOPs Comparison and Block Size
As long as I know - IOps is a IO operations per second on a given data block. But size of that given block if often omitted.
I assume, that block size does not make difference as long as its ...
0
votes
1
answer
36
views
Advantage of Logs and DBs on separate SAS HDDs on dedicated server?
Is there any significant performance advantage of having log data (application + DB) and DB data on separate SAS drives (for a dedicated server running CentOS) because writes of logging and db will ...
- 117
3
votes
1
answer
4k
views
KVM direct disk access vs raw files
I have to setup 14TB of raid storage space (which just shows up as /dev/sdb).
My question is:
Is there any impact on performance from a kvm guest if I create ext4 on this large drive directly, mount ...
- 31
1
vote
2
answers
673
views
Confused about Azure Storage Performance
In my experience so far with Azure the biggest bottleneck seems to be Storage performance. I did a few tests with Crystal Disk Mark and as you can see below the performance is very poor. Especially ...
- 171
2
votes
1
answer
420
views
How can I determine which files have high I/O in Windows?
In trying to reduce the storage bottleneck on a Windows server (2008 R2) I need to find the files that have the highest I/O over a long period of time (aka 'hot' files) so I can move those to SSD ...
- 398
1
vote
1
answer
172
views
HDD seq. read performance
I read the specs for one of the latest Hitachi HDD, https://www.hgst.com/sites/default/files/resources/USC10K1800_ds.pdf . It indicates a speed of
SAS 12 Gb/s
which is 1.5 GB/s. Is it the sequential ...
- 129
3
votes
2
answers
5k
views
Optimize Linux file system for reading ~500M small files
We're seeing performance problems on a file system that contains ~500 M files that are relatively small (typically tens of kb) and ~150 K directories. Files access is predominantly reads and writes ...
- 131
2
votes
1
answer
483
views
SAS direct dual link connection to an embedded LSI Fusion-MPT controller
This is a hardware question.
I have a dual-link SAS hard drive and a Intel S5000XVNSAS motherboard without the drive cage option. Recently, I bought a direct dual-link cable in amazon. But when ...
0
votes
2
answers
3k
views
RAID1 very slow with LSI card and 2x Samsung EVO 850 drives
not sure if I'm checking it correctly, but here is what I have.
Desktop, LSI 9260-8i card, see config below. 2x 500Gb Samsung 850 EVO drives.
I also have Samsung 830 drive attached to motherboard ...
- 213
0
votes
2
answers
198
views
Parallel disk seek with JBOD
I'm looking to build a 80 TB file based storage system. Ideally, I'd like to do this with a large JBOD and one server with a lot of cores.
My question is: Will JBOD containers (like Intel's) allow ...
19
votes
3
answers
5k
views
Is running permanently in a VMWare snapshot bad for performance?
I understand that the VMWare KB frowns upon long running snapshots mainly due to two things (In my opinion)
Taking tons of snapshots can fill up the data store. Snapshots are simply delta files. ...
2
votes
1
answer
120
views
VMware VMFS5 Local Storage LUN Sizing
We have a VM ESXi 5.x host using local storage (DAS).
It has 2 relatively large Logical Drives in RAID10 (10 HDD x1TB each). So, 2 LUNs of 5TB each.
We are considering creating just two datastores,...
- 393
10
votes
2
answers
24k
views
Slow disk performance on Azure Virtual Machine
Okay, so first of all, allow me to say that I'm not an operations person, but a developer. So I'm going into a bit of unknown land here, so please bear with me.
I would like to use a Azure Virtual ...
- 203
3
votes
2
answers
3k
views
Performance associated with storing millions of files on NTFS
Does anyone have a method / formula, etc that I could use - hopefully based on both current and projected numbers of files - to project the 'right' length of the split and the number of nested folders?...
- 15.6k
6
votes
1
answer
7k
views
Sizing of journal disks with Microsoft Storage Spaces parity volume
I'm currently speccing out a new backup server. It's going to be running Windows Server 2012 R2 with Direct Attached Storage, and I'm considering using Storage Spaces rather than using a RAID card. (...
- 2,625
-1
votes
1
answer
2k
views
Why does a low level format take very long time?
I can write ~100MB/s on a particular 4TB SAS drive when it have a file system on it, but a low level format on this drive took >16 hours. That's ~70kB/s. Since there is not filesystem and only zeroes ...
- 2,540
10
votes
2
answers
3k
views
Why is my RAID1 read access slower than write access?
I have done some simple performance tests and it seems that reading from my RAID1 is slower than writing:
root@dss0:~# for i in 1 2 3; do dd if=/dev/zero of=/dev/sda bs=1048576 count=131072; done
...
- 1,336
14
votes
7
answers
16k
views
How do storage IOPS change in response to disk capacity?
All other things being equal, how would a storage array's IOPS performance change if one used larger disks.
For example, take an array with 10 X 100GB disks.
Measure IOPS for sequential 256kb block ...
- 2,608
9
votes
3
answers
10k
views
Why would Linux VM in vSphere ESXi 5.5 show dramatically increased disk i/o latency?
I'm stumped and I hope someone else will recognize the symptoms of this problem.
Hardware: new Dell T110 II, dual-core Pentium G850 2.9 GHz, onboard SATA controller, one new 500 GB 7200 RPM cabled ...
- 689
4
votes
1
answer
2k
views
Mail Server Storage Backend: Filesystem VS Database
I am trying to compare mail server software to decide which to use at the ISP where I work. I'm confused mainly about which storage backend is better for such use (again, at an ISP, i.e expected large ...
- 141
5
votes
2
answers
2k
views
iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
The partition is in RAID1 using software RAID.
Command line used: iozone -s 190g -r 300k -O -b test_file
OPS Mode. Output is in operations per second.
...
- 599
0
votes
3
answers
363
views
Linux: Get information/stats about storage network
I want to know as much info as possible about the SAN a linux server is connected to. What commands will give me the most comprehensive view? Simply knowing the WWN of LUNS doesn't do much for me. ...
- 616
2
votes
2
answers
5k
views
HP Smart Array P822 w. Dual D2700 - cabling for best performance
Reading the D2600/D2700 User guide, there's a section called "Cabling examples", but none of the examples match what we intend to do.
I just want to make sure that it makes sense, before we do it.
...
- 118
5
votes
3
answers
6k
views
How many IOPS do I need? My workload bottleneck is storage
How can I know how many IOPS I need my storage to deliver for my overloaded Linux server?
I have a server and I know it has storage as its bottleneck. I would like the bottleneck not to be storage, ...
- 2,946
2
votes
1
answer
4k
views
interpreting IOPS these Bonnie++ and iostat results
I am looking to interpret the IOPS from a few runs of bonnie++ using iostat simultaneously and using the benchmarking from this post:
Can I determine IOPS on a disk array using bonnie++?
bonnie++ ...
- 29
0
votes
1
answer
179
views
How to benchmark a whole storage infrastructure
I'm a 100% sure that I'm not the first one to consider benchmarking a whole infrastructure, but yet I haven't found any relevant information on how to face this challenge.
As I guess we all know what ...
user294330
2
votes
0
answers
252
views
Infiniband link is at 4X SDR with DDR HCAs; has anyone seen this before?
I've recently upgraded my Infiniband HCAs from SDR to DDR, however upon upgrading, it has not improved the link speed at all. Receiving the following output from 'iblinkinfo':
CA: ip-san-1-au3a ...
- 1,462
2
votes
0
answers
1k
views
windows2012 failover cluster + storage. Considerations
We plan to build a failover storage with Windows2012r2. There will be 2 or 3 physical machines with 2 virtual machines on each server. One virtual machine building a failover, second one being storage ...
- 184